Omnidirectional Free-Space Optical (FSO) Receivers Gabriel A. Cap, Hakki H. Refai, and James J. Sluss, Jr. School of Electrical and Computer Engineering The University of Oklahoma – Tulsa Tulsa, Oklahoma USA

ABSTRACT

Free-space optics (FSO) is a technology that utilizes modulated light beam to transmit information through the atmosphere. Line-of-sight connection between both FSO transceivers is a necessary condition to maintain continuous exchange of voice, video, and data information. To date, the primary concentration of mobile FSO research and development has been toward the accurate aligning between two transceivers. This study introduces an advanced FSO receiver that provides wider receiving angle compared with that of conventional FSO systems. We present data from measurements of optical power, which were very promising, and indicated that these advanced FSO receivers are suitable for FSO alignment applications and perform favorably with similar FSO receivers.

I.

INTRODUCTION

Free-space optics (FSO) is a growing technology that is being used more and more in communication today. Due to its smaller size and cost effective solutions FSO could become the choice of service for broadband technology in the future. Research has progressed into the “last mile” network in order to connect users with high-speed internet service. This technology has many advantages over fiber optic communication for several reasons. Mainly, FSO can reach any area through the air without the expensive and inconvenient process of laying fiber optic cable in dense areas of the metropolitan area [1]. Free-space technology if used correctly could become the choice communication industry in the future. Currently, FSO is being used in many different ways. They are used as backup and redundant link to existing fiber optic networks to eliminate downtime and provide a secondary network for use. Currently bandwidth of up to 2.5 Gbps of data as well as other communications such as voice or video can be transferred through the air. This technology is still young and delicate, as very few software has been developed for comprehensive FSO design [2]. Some software exists towards signal propagation or optical system simulators, but as this industry expands system level design software for FSO links as well as simulation tools for FSO systems will be created and used to help increase the use of this technology [3]. In comparison to fiber optic networks, FSO exceeds fiber by a number of ways. Major advantages include the following: 1)Fewer licensing requirements; 2)high data rates; 3) smaller more compact size for easier use; and 4) there is no need to spend time and money digging for fiber under roads and buildings to provide broadband to different customers. [4] This cost effective technology could eliminate the need to run hard wire lines to companies that need fiber to their buildings to support their Local Area Networks (LANs) as well as provide a service that produce a faster bandwidth at cheaper rates. However, FSO can be complicated by several environmental as well as technical problems. Factors such as wind, small earthquakes, and even simple building sway can cause major problems. Other factors could also be temperature, pressure, humidity and even the wavelength for transmission. Recent studies have shown that atmospheric conditions and even sunlight have major control on the FSO feed [5]. The major effect that these conditions have on the FSO system is the loss of alignment and line of sight (LOS). If the LOS is blocked or the alignment is thrown off between the transceivers, the link could be broken and therefore must be realigned and

reconstructed to reconnect the system. If LOS is slightly lost, a realization of the loss power transferred in the system is also noticed. Currently, conditions that cause poor performance cause a decrease in power; therefore, more powerful and more expensive lasers must be used in order to carry out the necessary tasks. This is a major problem currently in FSO. The expense in keeping the transceivers online through any inclement weather or other factors is reason enough to stay with fiber optics until a less luxurious arrangement is conceived. In general, a system that could be created to keep alignment under any circumstance and furthermore any angle of transmission would be ideal as long as the power is kept nearly constant. With a similar transfer of power, the system could be kept online no matter what angle the transceiver used to converge on its target. In this paper, a system is developed using new techniques to send information from the transmitter to the receiver from any direction, a full 360°, while maintaining the same power broadcasted. If successful, this technique could be used during any situation, in good or poor weather, and alignment could be preserved and the FSO connection left intact. The remainder of this paper describes the theory, experimental setup, and results of this procedure, in preparation for the description of a new process in alignment of FSO systems.

Figure 1: Photodetector and Lens

II.

EXPERIMENTAL SETUP

This section describes the experimental methods used to investigate the effects of omnidirectional receivers for FSO systems. The laser source used was a green Helium-Neon laser beam transmitter. It was mounted and placed in front of a diverging convex lens in order to spread out the beam. The beam was diverged enough to cover the entire converging lens on the opposite end of the setup. The converging lens was then set to center the beam in the middle of the position sensing diode (PSD). The PSD was mounted in the focal plane of the converging lens. The PSD was then wired to an A/D converter and the data was taken and compiled by the computer.

Figure 2: The experimental setup was used for initial calibration and testing.

The beam was carefully aligned to the center of the PSD followed by placing the lens into position to create the same effect. The lens was plano-convex to allow the most beam divergence and convergence to the focal plane. This divergence was allowed in order to be able to cover the entire lens from an angle and direction.

Figure 3: Lens set up to transmit laser beam. Once these positions were setup, the converging lens and the PSD were no longer moved to allow for constancy in the experiment. The beam was then diverged from multiple angles and diverged on different points on the PSD in the focal plane.

PSD

Converging Lens

Ө Angle Laser Source

Laser Source

Laser Source Figure 4: Measurements taken from different angles.

By diverging the beam to cover the entire converging, concave lens, the laser would be focused onto a point in the focal plane on the PSD. Measurements were then taken concerning the power of the beam. Position measurements were taken for reference as well. Power measurements were taken from different horizontal and vertical angles. The laser beam source was mounted above and below the horizontal corresponding to the PSD to take measurements from angles simulating all directions. In total nine different angles were tested to verify the theory and validate the concept. Measurements and data were recorded in order to further validate the experiments. The recorded data is shown in table 1.

Photodetector

Figure 5: PSD receiving beams in focal plan from various angles

II.1. Experimental Procedures To accurately take the position and power measurements a strict procedure was followed. A power measurement was taken using a separate optical power meter for reference; also, a power measurement was taken using this optical power meter while the beam was diverged with the lens. The beam was centered on the PSD and a power measurement taken. The converging lens was placed so that the PSD was in the focal plane, with careful consideration to keep the same position of the beam and not to alter it as it went through the lens. After the converging lens was set, the diverging lens was placed also with careful consideration to keep the beam on the same spot on the PSD to accurately center the device. The first position and power measurements were then taken with the PSD accurately zeroed. The device was then moved accordingly to cover the other eight spots on the PSD. The laser beam and diverging lens were moved until the beam was set in the correct position correlating to the position being measured. Once the correct position was obtained, the power was also measured through the PSD. The dimensions of each setup were also measured including the height to the center of the laser, the length from lens to lens of each measurement. These results were then placed in an excel spreadsheet to be analyzed.

III.

EXPERIMENTAL RESULTS

The PSD displayed four output voltages corresponding to position and power of the beam. These outputs were taken and run through several calculations to accurately determine the position and power of the beam point. The four outputs were labeled as 1) Change Y, 2) Change X, 3) Sum Y, and 4) Sum X. These four outputs help determine the exact position and power. To accurately find the x, y coordinate position of the beam, the change of the coordinate was normalized by the sum of x and y. For example, to gain the Y coordinate the Change Y output was divided by the sum of the absolutes of sum x and sum y. In the same way, the X coordinate was found by dividing the Change X by the sum of the absolutes of sum x and sum y. Power was then calculated by taking the average of the sum X and sum Y, dividing by 10,000 to subsidize the gain, and finally dividing by .37 to factor in the responsivity corresponding to the specific wavelength [6]. Measurements were taken through a program written in Software for Visual Studios and all data was transferred to Microsoft Excel spreadsheets to run the calculations for the results. To accurately tests the theories, the experiments were setup to run with a series of different lens and focal lengths to understand how those affected the receiver. Lenses were used with diameters of 3cm, 5cm, and 8cm.

Focal lengths of 35mm, 50mm, 75mm, and 100mm were used. Tests were run gathering data at the nine positions corresponding to different angles. The results are shown below.

Position and Power for Each Lens at Specific Angles Lens Position 1) 0.0,0.0

X Position Y Position Power (mW)

Theoretical Values 0.0 0.0 1.35

5cm75mm

8cm100mm

7.5cm50mm

5.0cm35mm

3.0cm33mm

-0.01 0.00 0.99

0.02 0.01 0.88

0.01 0.05 1.08

-0.01 0.08 1.11

-0.08 -0.04 1.13

0.04 0.46 1.04

-0.01 -0.44 0.96

0.02 0.44 0.91

-0.02 -0.41 1.08

-0.01 -0.37 1.01

2) 0.0,-0.5

X Position Y Position Power

3) 0.0,0.5

X Position Y Position Power (mW)

0.0 0.5 1.35

-0.02 -0.44 0.79

0.05 0.48 0.94

-0.02 -0.40 0.88

0.03 0.41 1.05

0.01 0.41 0.99

4) -0.5,0.0

X Position Y Position Power (mW)

-0.5 0.0 1.35

-0.43 0.01 1.03

-0.42 0.06 0.94

-0.39 -0.10 0.96

-0.41 0.01 1.16

-0.39 -0.07 0.86

5) -0.5,-0.5

X Position Y Position Power (mW)

-0.5 -0.5 1.35

-0.43 -0.44 0.96

-0.43 0.43 0.99

-0.41 -0.41 0.99

-0.39 -0.33 1.00

-0.38 -0.34 0.99

6) -0.5,0.5

X Position Y Position Power (mW)

-0.5 0.5 1.35

-0.42 0.47 1.07

-0.42 -0.44 1.00

-0.39 0.43 0.93

-0.40 0.40 1.13

-0.36 0.40 0.94

7) 0.5,0.0

X Position Y Position Power (mW)

0.5 0.0 1.35

0.42 0.04 1.00

0.42 -0.01 0.96

0.42 0.02 0.94

0.40 0.05 1.02

0.38 -0.03 0.81

8) 0.5,-0.5

X Position Y Position Power (mW)

0.5 -0.5 1.35

0.42 -0.42 0.92

0.43 -0.43 0.95

0.41 -0.44 0.98

0.40 -0.31 1.04

0.40 -0.29 1.00

9) 0.5,0.5

X Position Y Position Power (mW)

0.5 0.5 1.35

0.40 0.47 1.14

0.41 0.43 0.92

0.43 0.39 0.95

0.40 0.41 1.10

0.40 0.39 0.96

Average Power

1.35

0.992

0.951

0.958

1.077

0.964

The main concern for the project was the average power of each lens at the correct at different positions along the PSD. Overall, the 5cm diameter with the 35mm focal length had the best average power, but also the 5cm with the 75mm focal length also had a good average power.

Average Power for Each Lens 1.35 1.40 1.077

0.992

1.20

0.964

0.958 0.951

1.00 Theoretical 5.0cm-35mm

0.80

5cm-75mm

Average Power (mW) 0.60

3.0cm-33mm 7.5cm-50mm

0.40

8cm-100mm

0.20 1 0.00 Theoretical

5.0cm-35mm

5cm-75mm 3.0cm-33mm Lens Types

7.5cm-50mm

8cm-100mm

The power did dissipate some due to traveling through two lenses, but the power did not decrease to an extent where a connection could be lost. The graph above shows that a 5cm diameter was optimal for retaining the most power, but the focal length did not play as big of a role as previously thought. As long as the PSD was aligned in the focal plane, the power and position could be transferred with very good precision. However, in some early trials, the PSD was not placed in the focal plane and the results quickly showed that power was lost as well as accurate positions could not be calculated. It was essential to converge the beam to a single point in order for the PSD to read and output efficiently. Although the power readings did not come with 10% of the actual reading this could have occurred for a number of reasons. The power may have been read incorrectly by the PSD, and to date no study has been found to quantify the loss of power while traveling through lenses. A study should be done to discover the power absorbed by plano-convex lenses.

IV.

CONCLUSION

These experiments show a new technique for the introduction of an advanced FSO receiver that provides a wider receiving angle compared with that of conventional FSO systems. By further expanding these tests, it is calculated that by using three lenses side by side, an entire 180-degree span can be covered and information received from. Furthermore, with the addition of only 2 more lenses for a total of five the entire 360 degree spectrum can be covered to allow for reception by the PSD from any angle regardless of alignment and connection site. Further study to improve this technique could show that LOS could be stabilized and allowed to be used under a number of situations including, adverse weather, building sway, and misalignment due to other natural factors. By eliminating these factors, connection from point to point, or point to multipoint could be maintained easier and more importantly more efficiently in order to fully supply the receiver with the information from the transmitter.

REFERENCES

[1] [2] [3] [4] [5] [6]

H. Wiollebarndand B. S. Ghuman, Fiber Optics Without Optics,” IEEE Spectrum 41-45 (August 2001). L. C. Andrews and R. L. Phillips, “Free space optical communication link and atmospheric effects: single aperture and arrays,” Proc. SPIE 5338, 265-275 (2004). G. Shaulov, J. Patel, B. Whitlock, P. Mena, and R. Scarmozzino, “Simulated-Assisted Design of Free Space Optical Transmission Systems,” RSoft Design Group, Ossining, NY. S Arnon, “Optimization of Urban Optical Wireless Communication Systems,” IEEE Trans. 1536-1276 (2003). M. Gebhart, E. Leitgeb, J. Bregenzer, “Atmospheric effects on Optical Wireless links,” ConTEL 2003. http://pacific-sensor.com/pdf/series-7.pdf